Re: [htdig3-dev] Re: iTools version of ht://Dig on MacOS X


Subject: Re: [htdig3-dev] Re: iTools version of ht://Dig on MacOS X
From: Geoff Hutchison (ghutchis@wso.williams.edu)
Date: Wed Feb 16 2000 - 12:12:02 PST


On Wed, 16 Feb 2000, Gilles Detillieux wrote:

> as well. It's a pretty simple patch, but not exactly clean, as they
> never bothered with declarations for the my*() functions and let the
> compiler assume int for everything. I think a cleaner approach would

Yes, I took a look last night and wasn't impressed. But then again,
they're just distributing binaries for the most part, so they don't have
to worry about how cleanly it compiles. I was clearly the first person to
ask since it took them a week or so to make it available.

(In their defense, this is only a very small part of their package, so I
doubt they spent any significant time on the modifications.)

> be to rename the method names that conflict with the socket library
> functions, so that you can do away with the "::" overrides, and add a

Yes, looking at their changes, I think this may be a good thing,
especially for portability. Should we just use the Ht prefix or try to go
with synonyms?

> test in the autoconf stuff for sys/malloc.h as well as malloc.h. Come
> to think of it, why use malloc.h at at all? DB2_db::db_init() uses
> calloc() to allocate and clear a DB_ENV object, but it should probably
> use "new" and memset() instead, then you can do away with malloc.h.

I'm pretty sure the GNU regex.c uses malloc. However, it seems to be
pretty sophisticated about working out which it can use.

-Geoff

------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
htdig3-dev-unsubscribe@htdig.org
You will receive a message to confirm this.



This archive was generated by hypermail 2b28 : Wed Feb 16 2000 - 12:15:07 PST